Given the unpleasant role that its aluminium business has played in Rio Tinto’s recent history, Sam Walsh could be forgiven for trying to pretend it doesn’t exist. Instead, however, he showcased it at last night’s investor briefing in London and outlined plans to invest the best part of $US1 billion in expanding it.
The debt-financed $US38bn acquisition of Alcan just ahead of the financial crisis brought Rio to its knees, nearly cost it its independence and resulted in around $US30bn of writedowns. That blight on Rio’s record is, of course, history and a lot has changed within the sector since Rio absorbed those massive impairments.
Rio itself has been part of the dramatic and traumatic structural changes that have been made to the sector, selling, reducing or closing about a quarter of its aluminium capacity and a third of its alumina capacity since the crisis as well as carving into the cost base of the continuing operations.
The remaking of the sector has been industry wide. Over the post-crisis period, for instance, Alcoa has shut down about a million tonnes of aluminium production capacity while also reducing the capacity of its continuing operations by the best part of another million tonnes. Rusal has taken more than 15 per cent of its capacity out of the market.
It has taken time for the effects of the large-scale cut-backs in production to impact the market because China, which represents about half the global production base, kept increasing its output despite the over-supplied position of the market and depressed prices, which meant about half its capacity was loss-making at a cash level. More recently, however, it has been closing its higher-cost capacity as China’s leadership has increasingly focused on driving out sub-economic activity.
The industry appears to have bottomed out, supply and demand are broadly balanced and the price has firmed to around the $US2000 a tonne level at which the industry as a whole breaks even.
Rio, which has shifted most of its production into the first quartile in terms of costs, generated $US1.1bn of earnings before interest, tax, depreciation and amortisation in the first half of this year from its aluminium division and has become increasingly confident about the outlook.
In last night’s presentation it forecast four per cent compound annual growth in demand for primary metal over the next decade, with continuing shortfalls in supply pushing up the price over the next five years. That does, of course, imply discipline from producers that do have significant latent capacity they could bring back on stream.
Rio has a particularly strong position in bauxite, where it is the world’s leading producer. Walsh said that, assuming consensus prices over the next five years, his bauxite business was expected to generate EBITDA margins of more than 50 per cent. Demand for bauxite was forecast to grow at 8 per cent per annum over the next decade.
That explains why Rio is now "prioritising" its South of Embley phased 50 Mtpa expansion project at Weipa in Queensland, with a decision on the proposed investment to be made next year. Rio describes the project as a 'tier one' opportunity.
Rio, which has Glencore’s Ivan Glasenberg breathing down its neck, needs its aluminium and copper businesses to make a larger contribution to group earnings now completely dominated by its fabulous iron ore business.
In the near term, some production issues in copper will constrain its contribution, although Rio does have some very big and attractive longer term growth projects in its portfolio.
That, and the dive in iron ore prices, makes a continuing surge in the earnings from the aluminium division that much more significant for Rio in the short to medium term.
While the Alcan acquisition will forever remain an horrendously expensive and value-destructive blight on Rio’s history, the written-down and radically restructured business is leveraged to an improvement in prices and does have the capacity to make a very material contribution to the group’s results. Walsh appears increasingly confident that it will.